This is a study of the natural history of pathogenic events leading to the clinical onset of insulin dependent diabetes (IDD). The goals are to gain insights into the disease process so as to be able to predict who will develop the disease and when. The three study populations are: high risk relatives of IDD probands, intermediate risk patients with autoimmune thyroid and/or adrenal diseases and their families, and low risk general population of school children. The three University Medical Centers in Florida as well as 5 satellite groups will participate. Current predictive markers include islet cell (cytoplasmic) autoantibodies (ICA), insulin autoantibodies (IAA), elevations of fasting plasma glucose (more than 110mg/ml) and impaired early phase insulin release to IVGTT, with risks determined by life table analyses of data obtained from single observations. We now intend to estimate risks from sequential data obtained through multiple observations and the recursive partitioning of risk groups, and to add several new potential immunologic and metabolic markers. Autoantibodies to an islet cell protein of 64KDa (64KA) may be the best early predictive marker. The 64KDa antigen has been shown recently to be the 65KDa variant of glutamic acid decarboxylase (GAD). This antibody will be identified by an immunoassay utilizing recombinant GAD proteins. Autoantibodies to insulin receptors and lymphocyte antigens will also be sought. Defects in cellular mediated immunity have also been identified in relatives with impending IDD. These defects include relative lymphopenia, increased circulating T cells bearing gamma-delta T cell receptors, decreased T cell helper-inducer function and persistence of infant frequencies of circulating CD5+ B cells into adulthood. Metabolic parameters to be explored include possible hyperglucagonemia and insulin resistance measures made through infusions of stable label glucose and minimal modelling analyses. These continued studies will provide an increasingly accurate basis for predicting the risk of IDD, a necessity in the development of intervention strategies to prevent IDD.